English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Significant performance degradation of automatic speech recognition (ASR)systems is observed when the audio signal contains cross-talk. One of therecently proposed approaches to solve the problem of multi-speaker ASR is thedeep clustering (DPCL) approach. Combining DPCL with a state-of-the-art hybridacoustic model, we obtain a word error rate (WER) of 16.5 % on the commonlyused wsj0-2mix dataset, which is the best performance reported thus far to thebest of our knowledge. The wsj0-2mix dataset contains simulated cross-talkwhere the speech of multiple speakers overlaps for almost the entire utterance.In a more realistic ASR scenario the audio signal contains significant portionsof single-speaker speech and only part of the signal contains speech ofmultiple competing speakers. This paper investigates obstacles of applying DPCLas a preprocessing method for ASR in such a scenario of sparsely overlappingspeech. To this end we present a data simulation approach, closely related tothe wsj0-2mix dataset, generating sparsely overlapping speech datasets ofarbitrary overlap ratio. The analysis of applying DPCL to sparsely overlappingspeech is an important interim step between the fully overlapping datasets likewsj0-2mix and more realistic ASR datasets, such as CHiME-5 or AMI.

Analysis of Deep Clustering as Preprocessing for Automatic Speech Recognition of Sparsely Overlapping Speech